And I’d disagree with both.
Some browsers, like the one you should be using, have anti-fingerprinting tech in them.
Anti-fingerprinting tech just produces a different fingerprint. Google knows e.g. when things are scrambled but certain other things stay the same.
You can try using a different device but even then, I occasionally get recommending things that are definitely influenced by my roommates (i.e. on the same WiFi)
Using something that prevents fingerprinting helps, but only if you don’t use that browser all the time — otherwise it’s just another fingerprint — and still on the same network.
> A programming flaw in its cloud services also allowed China-backed hackers to steal email from federal officials. On Friday, Microsoft said it would stop using China-based engineers to support Defense Department cloud-computing programs after a report by investigative outlet ProPublica revealed the practice, prompting Defense Secretary Pete Hegseth to order a review of Pentagon cloud deals.
Literally wasn’t the right word. Maybe control just needs more context, like “control of your usage”?
You are taking the “no control” thing too literally.
Its fine if you just rebase at the end manually, but not good if you don't, your history will be cluttered and as hard to read as the codebase.
Eventually most people who use coding tools will have low knowledge of what is being generated and then they probably never rebase either...
I just commit with a “wip!”-prefaced message whenever the LLM pauses and says it’s finished, including new files. You can squash and cleanup later, or revert back to a state before it screwed up.
Also doubles as a way to cohesively look at the changes it made without all the natural language and recursive error/type fixing it does while working.
I don’t understand why people are making it so complicated. You’re saving a minute per iteration with the LLM, tops, at risk of losing control or introducing hard to find issues. It is the definition of diminishing returns.